Automation retainer

Stop losing engineering days to releases

Your release pipeline and repetitive engineering work are owned and monitored by someone who is on the hook when they break — not just built once and handed back.

from$2,000/mo

Book a discovery call

$2,000/mo essentials · $5,000/mo full · $150–300/hr ad-hoc
Monthly, 3-month minimum · 30 days' notice after that

The recurring cost

None of this is a project

A release takes a senior engineer most of a day, every time.
CI broke on a dependency nobody chose to upgrade.
Store submissions get postponed because nobody wants to own them.
Monitoring exists, and nobody has read it since it was set up.

It is work that arrives every month and never finishes, which is why it loses to whatever is on the roadmap. A build ends. A pipeline does not — so it needs an owner, not a project.

Included

What the monthly fee actually buys

EAS build and store-submit pipelines, maintained and monitored.
OTA update lanes for JS-only changes — shipped without a store-review wait.
Cron jobs, monitoring, and alerting owned end to end.
Agency overflow: white-label RN/Expo delivery under your brand, NDA-clean.

Most agencies screw things to the surface and hand you the screwdriver. We go through, fasten on the far side, and stay on the hook when it moves.

Proof

We run this on our own releases

The automation we sell is the automation this repo runs on — GitHub Actions workflows for quality gates, EAS builds and store submission, OTA update lanes, releases, and nightly store screenshots. We screen-share the whole matrix on the first call.

That is the whole case, and it is deliberately the only one on this page. There are no logos or testimonials here: evidence you can check is worth more than evidence you have to take on trust, so what you get instead is the system itself, on screen, on the first call.

What we can do instead is show you the work. On 1 August 2026 we found that our own CI had not run in months — every push and pull request gate was filtered to a branch this repository has never had, so the checks were passing by never executing, and one publish workflow had been rejected outright on every push since July. We found it, fixed it, and wrote down which of the remaining workflows we would not switch on without a decision from the owner. That is the job, and it is the exact failure mode a retainer exists to catch.

We will walk you through all of it on the first call — the workflow matrix, a real run, the OTA lane, and that fix — in the repository they run in.

Before you book

The questions that decide it

Why is this so much cheaper than an agency?: Because the cost structure is different, not the engineering. A traditional agency price carries account management, bench time, and a project team that rotates — you are paying for the org chart as well as the work. We price the work. The stack is also one we run in production for our own products, so the scaffolding every project needs — auth, subscriptions, CI, release automation, store submission — is not rebuilt from zero at your expense.
What happens if you become unavailable?: Continuity is designed in rather than promised. Source code and infrastructure live in your accounts from day one, not ours — you can revoke our access this afternoon and keep shipping. The pipeline is configuration in your repository, not knowledge in somebody's head: the build lanes, the OTA lane, the store submission and the monitoring are all committed, reviewable, and runnable without us. Retainers run month to month after the three-month minimum, either side can end with 30 days' notice, and ending one includes a documented handover rather than a cliff. No vendor of any size can honestly guarantee availability, so the question worth asking any of them is what you are left holding if it stops. Here it is everything.
How does payment work?: Retainers are invoiced monthly in advance, on the same date each month, against a three-month minimum and rolling monthly after that. Fixed-scope builds are typically 50% to start and 50% on delivery, with larger builds split across milestones instead. Either way the exact structure is written into the scope document before any work begins.
What does the three-month retainer minimum mean?: It means the first three months are committed, because a release pipeline cannot be taken over and stabilised in four weeks — the first month is largely learning your systems, and cancelling at that point leaves you worse off than before. After month three it runs month to month and you can end it with 30 days' notice. No annual lock-in, no auto-escalating rate.
What if a quiet month goes by and nothing breaks?: That is the outcome you are paying for, and it is a fair question to ask before signing. A retainer buys ownership and availability, not a ticket quota: releases still ship, dependencies and OS versions still move underneath you, and monitoring still needs someone reading it. If genuinely quiet months repeat, the honest answer is to move you to the ad-hoc hourly rate rather than keep billing a retainer you have outgrown — we would rather lose the line item than have you resent it.
Do I need a retainer or a fixed-scope build?: A build if the thing does not exist yet and has a defined finish line. A retainer if it already ships and the cost is recurring — releases, CI breakage, store submissions, upgrades. Plenty of engagements start as a build and become a retainer at launch, which is the point at which the recurring work actually begins. If you are unsure, the discovery call is the cheapest way to find out, and a straight 'you do not need us monthly' is a normal outcome.
Who owns the code?: You do, outright, at handover — source and infrastructure in your own accounts. No license, no hosting lock-in, no per-seat fee afterwards.

Prices and terms on this page are informational and are fixed per engagement in a written scope. They are an invitation to start a conversation, not a binding offer — see the terms. Full rate card, including fixed-scope builds, on pricing.

Worth thirty minutes?

Bring your release process and what it currently costs you. You leave with a straight answer on fit and a number — including “you do not need us monthly”, which is a normal outcome. Prefer email? arberrexha.dev@gmail.com.

Book a discovery call or just send a one-line email

We reply within one business day.

Stop losing engineering days to releases

Your release pipeline and repetitive engineering work are owned and monitored by someone who is on the hook when they break — not just built once and handed back.

from$2,000/mo

Book a discovery call

$2,000/mo essentials · $5,000/mo full · $150–300/hr ad-hoc
Monthly, 3-month minimum · 30 days' notice after that

What the monthly fee actually buys

EAS build and store-submit pipelines, maintained and monitored.

OTA update lanes for JS-only changes — shipped without a store-review wait.

Cron jobs, monitoring, and alerting owned end to end.

Agency overflow: white-label RN/Expo delivery under your brand, NDA-clean.

Most agencies screw things to the surface and hand you the screwdriver. We go through, fasten on the far side, and stay on the hook when it moves.

We run this on our own releases

We will walk you through all of it on the first call — the workflow matrix, a real run, the OTA lane, and that fix — in the repository they run in.

The questions that decide it

Why is this so much cheaper than an agency?

Because the cost structure is different, not the engineering. A traditional agency price carries account management, bench time, and a project team that rotates — you are paying for the org chart as well as the work. We price the work. The stack is also one we run in production for our own products, so the scaffolding every project needs — auth, subscriptions, CI, release automation, store submission — is not rebuilt from zero at your expense.

What happens if you become unavailable?

Continuity is designed in rather than promised. Source code and infrastructure live in your accounts from day one, not ours — you can revoke our access this afternoon and keep shipping. The pipeline is configuration in your repository, not knowledge in somebody's head: the build lanes, the OTA lane, the store submission and the monitoring are all committed, reviewable, and runnable without us. Retainers run month to month after the three-month minimum, either side can end with 30 days' notice, and ending one includes a documented handover rather than a cliff. No vendor of any size can honestly guarantee availability, so the question worth asking any of them is what you are left holding if it stops. Here it is everything.

How does payment work?

Retainers are invoiced monthly in advance, on the same date each month, against a three-month minimum and rolling monthly after that. Fixed-scope builds are typically 50% to start and 50% on delivery, with larger builds split across milestones instead. Either way the exact structure is written into the scope document before any work begins.

What does the three-month retainer minimum mean?

It means the first three months are committed, because a release pipeline cannot be taken over and stabilised in four weeks — the first month is largely learning your systems, and cancelling at that point leaves you worse off than before. After month three it runs month to month and you can end it with 30 days' notice. No annual lock-in, no auto-escalating rate.

What if a quiet month goes by and nothing breaks?

That is the outcome you are paying for, and it is a fair question to ask before signing. A retainer buys ownership and availability, not a ticket quota: releases still ship, dependencies and OS versions still move underneath you, and monitoring still needs someone reading it. If genuinely quiet months repeat, the honest answer is to move you to the ad-hoc hourly rate rather than keep billing a retainer you have outgrown — we would rather lose the line item than have you resent it.

Do I need a retainer or a fixed-scope build?

A build if the thing does not exist yet and has a defined finish line. A retainer if it already ships and the cost is recurring — releases, CI breakage, store submissions, upgrades. Plenty of engagements start as a build and become a retainer at launch, which is the point at which the recurring work actually begins. If you are unsure, the discovery call is the cheapest way to find out, and a straight 'you do not need us monthly' is a normal outcome.

Who owns the code?

You do, outright, at handover — source and infrastructure in your own accounts. No license, no hosting lock-in, no per-seat fee afterwards.

Worth thirty minutes?

We reply within one business day.